k-means clustering is a method of vector quantization, that aims to partition data points / observations into distinct groups based on their similarity with each other. The number of distinct groups is determined by the value k. The similarity of each group is represented by means of the data points.
This method is part of the unsupervised learning which help us find meaningful insight from the unlabeled data. This method can be powerful for e.g. customer segmentation, where companies would like to get insights on customers shopping behaviour on their product.
The K-Means Clustering algorithm can be described as per below graph. Credit is given to https://data-science-infinity.teachable.com/
This Notebook describes how to use the K-Means clustering method to calculate the supermarket customer segmentation. The results of this analysis will help the marketing team to tailor the marketing campaign of the new products to the customers' needs.